arXiv subjects

Wang Qi

Publications and source records attributed to Wang Qi.

8 recordsLinked to original sources

arXiv2025

Liaohe-CobotMagic-PnP: an Imitation Learning Dataset of Intelligent Robot for Industrial Applications

In Industry 4.0 applications, dynamic environmental interference induces highly nonlinear and strongly coupled interactions between the environmental state and robotic behavior. Effectively representing dynamic environmental states through multimodal sensor data fusion remains a critical challenge in current robotic datasets. To address this, an industrial-grade multimodal interference dataset is presented, designed for robotic perception and control under complex conditions. The dataset integrates multi-dimensional interference features including size, color, and lighting variations, and employs high-precision sensors to synchronously collect visual, torque, and joint-state measurements. Scenarios with geometric similarity exceeding 85\% and standardized lighting gradients are included to ensure real-world representativeness. Microsecond-level time-synchronization and vibration-resistant data acquisition protocols, implemented via the Robot Operating System (ROS), guarantee temporal and operational fidelity. Experimental results demonstrate that the dataset enhances model validation robustness and improves robotic operational stability in dynamic, interference-rich environments. The dataset is publicly available at:https://modelscope.cn/datasets/Liaoh_LAB/Liaohe-CobotMagic-PnP.

cs.RO↗

arXiv2022

TegFormer: Topic-to-Essay Generation with Good Topic Coverage and High Text Coherence

Creating an essay based on a few given topics is a challenging NLP task. Although several effective methods for this problem, topic-to-essay generation, have appeared recently, there is still much room for improvement, especially in terms of the coverage of the given topics and the coherence of the generated text. In this paper, we propose a novel approach called TegFormer which utilizes the Transformer architecture where the encoder is enriched with domain-specific contexts while the decoder is enhanced by a large-scale pre-trained language model. Specifically, a \emph{Topic-Extension} layer capturing the interaction between the given topics and their domain-specific contexts is plugged into the encoder. Since the given topics are usually concise and sparse, such an additional layer can bring more topic-related semantics in to facilitate the subsequent natural language generation. Moreover, an \emph{Embedding-Fusion} module that combines the domain-specific word embeddings learnt from the given corpus and the general-purpose word embeddings provided by a GPT-2 model pre-trained on massive text data is integrated into the decoder. Since GPT-2 is at a much larger scale, it contains a lot more implicit linguistic knowledge which would help the decoder to produce more grammatical and readable text. Extensive experiments have shown that the pieces of text generated by TegFormer have better topic coverage and higher text coherence than those from SOTA topic-to-essay techniques, according to automatic and human evaluations. As revealed by ablation studies, both the Topic-Extension layer and the Embedding-Fusion module contribute substantially to TegFormer's performance advantage.

cs.CL↗

arXiv2022

Parameter-Efficient Tuning on Layer Normalization for Pre-trained Language Models

Conventional fine-tuning encounters increasing difficulties given the size of current Pre-trained Language Models, which makes parameter-efficient tuning become the focal point of frontier research. Previous methods in this field add tunable adapters into MHA or/and FFN of Transformer blocks to enable PLMs achieve transferability. However, as an important part of Transformer architecture, the power of layer normalization for parameter-efficent tuning is ignored. In this paper, we first propose LN-tuning, by tuning the gain and bias term of Layer Normalization module with only 0.03\% parameters, which is of high time-efficency and significantly superior to baselines which are less than 0.1\% tunable parameters. Further, we study the unified framework of combining LN-tuning with previous ones and we find that: (1) the unified framework of combining prefix-tuning, the adapter-based method working on MHA, and LN-tuning achieves SOTA performance. (2) unified framework which tunes MHA and LayerNorm simultaneously can get performance improvement but those which tune FFN and LayerNorm simultaneous will cause performance decrease. Ablation study validates LN-tuning is of no abundant parameters and gives a further understanding of it.

cs.CL↗

arXiv2007

Quantum-classical transition for an analog of double-slit experiment in complex collisions: Dynamical decoherence in quantum many-body systems

We study coherent superpositions of clockwise and anti-clockwise rotating intermediate complexes with overlapping resonances formed in bimolecular chemical reactions. Disintegration of such complexes represents an analog of famous double-slit experiment. The time for disappearance of the interference fringes is estimated from heuristic arguments related to fingerprints of chaotic dynamics of a classical counterpart of the coherently rotating complex. Validity of this estimate is confirmed numerically for the H+D$_2$ chemical reaction. Thus we demonstrate the quantum--classical transition in temporal behavior of highly excited quantum many-body systems in the absence of external noise and coupling to an environment.

quant-ph↗

arXiv2007

Slow cross-symmetry phase relaxation in complex collisions

We discuss the effect of slow phase relaxation and the spin off-diagonal $S$-matrix correlations on the cross section energy oscillations and the time evolution of the highly excited intermediate systems formed in complex collisions. Such deformed intermediate complexes with strongly overlapping resonances can be formed in heavy ion collisions, bimolecular chemical reactions and atomic cluster collisions. The effects of quasiperiodic energy dependence of the cross sections, coherent rotation of the hyperdeformed $\simeq (3:1)$ intermediate complex, Schrödinger cat states and quantum-classical transition are studied for $^{24}$Mg+$^{28}$Si heavy ion scattering.

quant-ph↗

arXiv2006

Effect of phase relaxation on quantum superpositions in complex collisions

We study the effect of phase relaxation on coherent superpositions of rotating clockwise and anticlockwise wave packets in the regime of strongly overlapping resonances of the intermediate complex. Such highly excited deformed complexes may be created in binary collisions of heavy ions, molecules and atomic clusters. It is shown that phase relaxation leads to a reduction of the interference fringes, thus mimicking the effect of decoherence. This reduction is crucial for the determination of the phase--relaxation width from the data on the excitation function oscillations in heavy--ion collisions and bimolecular chemical reactions. The difference between the effects of phase relaxation and decoherence is discussed.

quant-ph↗

arXiv2004

Effect of a finite-time resolution on Schrödinger Cat States in complex collisions

We study the effect of finite time resolution on coherent superpositions of rotating clockwise and anticlockwise wave packets in the regime of strongly overlapping resonances of the intermediate complex. Such highly excited deformed complexes may be created in binary collisions of heavy-ions, molecules and atomic clusters. It is shown that time averaging reduces the interference fringes acting effectively as dephasing. We propose a simple estimate of the ``critical'' time averaging interval. For the time averaging intervals bigger than the critical one the interference fringes wash out appreciably. This is confirmed numerically. We evaluate minimal energy intervals for measurements of the excitation functions needed to observe the Schrödinger Cat States. These should be easily observable in heavy-ion scattering. Such an experiment is suggested.

nucl-th↗

arXiv2003

Experimental test of spontaneous correlation and anomalous sensitivity in finite highly excited many-body systems

We have tested recent suggestion of anomalous sensitivity in highly excited quantum many-body systems. Two independent measurements of cross sections for the 19F+93Nb strongly dissipative heavy-ion collisions have been performed at incident energies from 102 to 108 MeV in steps of 250 keV.In the two measurements we used different, independently prepared, 93Nb target foils with nominally the same thickness.The data indicate statistically significant non-reproducibility of the energy oscillating yields in the two measurements. The observed non-reproducibility is consistent with recent theoretical arguments on spontaneous correlation, slow phase randomization and chaos in highly excited complex quantum systems.

nucl-ex↗